REVIEW ARTICLE

Cardiology Journal 2011, Vol. 18, No. 1, pp. 817 Copyright 2011 Via Medica ISSN 18975593

Prognostic significance of QRS duration and morphology

Andrew Brenyo, Wojciech Zarba
Cardiology Division, Department of Medicine, University of Rochester Medical Center, Rochester, NY, USA

Abstract QRS duration and morphology, evaluated via a standard 12-lead electrocardiogram (ECG), represent an opportunity to derive useful prognostic information regarding the risk of subsequent cardiac events or therapeutic outcomes. Prolonged QRS duration, and the presence of intraventricular conduction abnormalities, usually indicate the presence of changes in the myocardium due to underlying heart disease. Prolonged QRS duration is often associated with depressed ejection fraction or enlarged left ventricular volumes, but several studies have demonstrated that this simple ECG measure provides independent prognostic value, after adjusting for relevant clinical covariates. Post-infarction patients with prolonged QRS duration have a significantly increased risk of mortality, although data associating QRS prolongation specifically with sudden death is less supportive. In non-ischemic cardiomyopathy, there is no evidence that QRS duration has prognostic significance in predicting mortality or sudden death. Prolonged QRS duration, and especially presence of left bundle branch block, seems to predict a benefit from cardiac resynchronization therapy in both ischemic and non-ischemic cardiomyopathy patients. Therefore, QRS duration and morphology should not only be considered a predictor of death or sudden death in patients after myocardial infarction, and in those suspected of coronary artery disease, but also as a predictor of benefit from cardiac resynchronization therapy in patients with heart failure, whether of an ischemic or non-ischemic origin. (Cardiol J 2011; 18, 1: 817) Key words: QRS duration, left bundle branch block, prognosis, mortality

Introduction
QRS duration (QRSd) and morphology, evaluated using a standard 12-lead electrocardiogram (ECG), have always been considered important prognostic markers, even in individuals without structural heart disease. The Manitoba study [1] consisted of a cohort of 3,983 men with a mean age at entry of 30 years, followed with regular examinations including ECG from 1948. During the 30 year observation period,

70 cases of sudden death occurred in men without previous clinical manifestations of heart disease. Left bundle branch block (LBBB) was found to be associated with a relative 13.8-fold increased risk of sudden death in this cohort of healthy men. Studies that are more relevant to everyday clinical practice typically involve patients referred for evaluation due to suspected coronary disease. Schinkel et al. [2] analyzed a large cohort of 1,227 patients referred for dobutamine stress echocardiography to evaluate suspected myocardial ischemia.

Address for correspondence: Wojciech Zarba, MD, PhD, Cardiology Division, Department of Medicine, University of Rochester Medical Center, Heart Research, Box 653, 601 Elmwood Ave., Rochester, NY 14642, USA, tel: 585 275 5391, fax: 585 273 5283, e-mail: wojciech_zareba@urmc.rochester.edu

www.cardiologyjournal.org

Andrew Brenyo, Wojciech Zarba, Prognostic significance of QRS duration and morphology

During a mean follow-up of more than four years, they found that patients with QRS duration 120 ms had twice the cardiac mortality as did patients with QRS < 120 ms. Multivariate analysis adjusted for age, gender, smoking, hypertension, diabetes, hypercholesterolemia, ST segment depression, angina, and abnormal stress echocardiogram showed that QRS duration 120 ms was significantly associated with mortality with a hazard ratio of 1.8. In another similar study of 4,033 patients with known or suspected coronary artery disease, Elhendy et al. [3] also demonstrated that QRS prolongation was associated with increased mortality after adjusting for relevant clinical covariates. These large studies, and other smaller reports, indicate that QRS duration and morphology is of major prognostic importance even in individuals without organic heart disease. Abnormal QRS duration and morphology frequently identify subjects with clinically undetected cardiac abnormalities and an increased risk of mortality.

Post myocardial infarction population

The standard 12-lead ECG should always be considered when performing risk stratification of patients who have experienced myocardial infarction (MI). Heart rate, atrioventricular conduction, fascicular block, bundle branch block (BBB) and the degree of ST-segment deviation have all been shown to influence survival after MI [4]. Most of the early work (before the era of percutaneous transluminal intervention [PCI]) focused on the relationship between BBB and prognosis post MI. Those early studies consistently associated the presence of BBB with increased short- and longterm mortality with estimates of 2030% at one year [57]. In more recent datasets, it has become clear that specific quantification of the QRSd, and not just the presence or absence of BBB (Table 1), can provide more prognostic information, such as the likelihood of 30 day mortality or sudden cardiac death (SCD) post MI [8]. Abnormal intraventricular conduction has been extensively examined as a risk stratification tool in diverse cardiovascular disease (CVD) populations, including patients with abnormal left ventricular (LV) function, dyssynchrony and those at increased risk of SCD. Most of this work has focused on patients with chronic CVD and will be discussed in detail below. The prognostic significance of quantitative QRS conduction post acute MI has been relatively under-studied in comparison with these more chronic forms of CVD. Existing data is main-

ly derived from post-hoc analysis of multi-center registries focusing on revascularization, with 30 day all-cause mortality as a common endpoint. Recently, Wong et al. [9] examined the relationship between QRSd and 30 day mortality post MI in a prespecified post hoc analysis of the Hirulog and Early Reperfusion or Occlusion (HERO)-2 trial. They examined QRSd at baseline and 60 minutes after fibrinolysis in over 12,000 patients. Only patients with normal QRSd (< 125 ms) or RBBB at both time points were included in the analysis. Patients who developed LBBB or QRS delay not meeting the criteria for RBBB were excluded. Any prolongation of the QRS within this studied population was associated with increased 30 day mortality, but only in patients with anterior MI, and not with inferior MI. In the anterior MI population, the authors noted a 3040% increase in 30 day mortality for every 20 ms increase in the QRSd from baseline. A similar relationship was found in the GUSTO-1 [8] population with QRS prolongation within the normal range (100 vs 80 ms) post anterior MI carrying the strongest association and greatest prognostic significance for 30 day mortality (OR 1.55; 95% CI 1.431.68). Shortcomings of this robust dataset include its compilation prior to the current PCI era, and the lack of data regarding the actual cause of death. Recent efforts have sought to overcome these limitations and describe the prognostic weight of QRS prolongation, as well as specify the cause of death, with particular attention paid to SCD. Bauer et al. [10] prospectively analyzed 1,455 survivors of acute MI aged 75 or under with a dichotomized QRS variable cutoff of 120 ms, and found a strong association of QRS prolongation with total mortality (HR 4.0; 95% CI 2.36.9 after multivariate analysis). However, the authors did not find a significant association with SCD. A decreased incidence of BBB was found in patients in this population when compared to studies performed in the pre-PCI era (6 vs 1020%) but with similar one year all-cause mortality (18.6 vs 20.8%) [5]. The study pointed towards patients with abnormal QRSd carrying higher post MI mortality, but not from SCD. In another study by Guerrero et al. [11], analyses of 3,053 post-infarction patients who underwent primary angioplasty showed that LBBB was associated with five-fold higher in-hospital mortality; the authors did not report prognostic value of QRS duration. In an effort to assess the efficacy of treatment with valsartan and captopril, and their combination, in high-risk patients after MI (The Valsartan In Acute Myocardial Infarction [VALIANT] trial [12])

www.cardiologyjournal.org

10
Population Continuous 100 vs 80 ms 30 day all-cause mortality QRS variable Primary endpoint Results (95% CI) OR 1.55 (1.431.68); p < 0.001 for death with anterior MI and 1.08 (1.031.13); p < 0.001 for any other infarct location HR for primary endpoint 4.0 (2.36.9); p < 0.001; HR for cardiac mortality 3.9 (1.97.8); p < 0.001; HR for SCD not significant QRS duration of > 105 ms highest risk for mortality; HR 8.49 (4.416.34); p < 0.0001 No association between QRS duration and risk of all-cause mortality or SCD HR for primary endpoint of 1.31 (1.061.64); 1.57 (1.032.4) and 1.31 (1.032.4) for each QRS duration quartile All-cause mortality with ten year follow up 30 day all-cause mortality 6% increase in death for each 10 ms increase in QRS duration; HR 1.06 (1.041.09); p < 0.0001 3040% increase in death in anterior MI only for every 20 ms increase in QRS duration; OR 1.42 (1.331.51); p < 0.0001 Dichotomous with 120 ms cut point All-cause mortality at two years Continuous without defined increment Dichotomous with 120 ms cut point Continuous with quartiles starting at < 75 to > 108 ms Continuous 20 ms increments starting at 80 ms Continuous 20 ms increments starting at 80 ms All-cause mortality at five years All-cause mortality at 2.5 years with SCD as a secondary endpoint Combined endpoint of HF, SCD, CD, stroke and MI at two year follow-up

Table 1. Prognostic significance of QRS duration in post-infarction randomized clinical trials.

Trial

GUSTO-1

34,166

STEMI and no confounding ECG factors at entry (paced rhythms, ventricular rhythms, or LBBB)

Cardiology Journal 2011, Vol. 18, No. 1

Bauer et al.*

1,455

Survivors of AMI under 76 years of age without a paced rhythm

Elhendy et al.

4,033

Patients with known or suspected CAD

DINAMIT*

674

Patients 640 days post AMI randomized to ICD vs standard therapy

www.cardiologyjournal.org

VALIANT*

403

Post AMI with LV dysfunction symptomatic HF

DIAMOND

1,510

Post AMI with LV dysfunction and without HF

HERO 2

12,456

STEMI with normal QRS duration (80120 ms) or RBBB at baseline

*Only trials to assess sudden cardiac death incidence and QRS duration post acute myocardial infarction; STEMI ST segment elevation myocardial infarction; ECG electrocardiogram; LBBB left bundle branch block; AMI acute myocardial infarction; CAD coronary artery disease; ICD implantable cardioverter-defibrillator; LV left ventricle; HF heart failure; RBBB right bundle branch block; SCD sudden cardiac death; CD cardiac death; CI confidence interval; OR odds ratio; HR hazard ratio

Andrew Brenyo, Wojciech Zarba, Prognostic significance of QRS duration and morphology

the authors included QRS duration as a prespecified subgroup. The primary outcomes included allcause and cardiovascular mortality as well as SCD events. Patients with QRS prolongation were found to have increased ventricular volumes, decreased left ventricular ejection fraction (LVEF), and a higher incidence of heart failure (HF), cardiovascular death and SCD. As observed in the HERO-2 and GUSTO-1 trials, the increased risk was seen even with QRS prolongation within the normal range. Subsequent work by Solomon et al. [13] with the VALIANT cohort found that risk of sudden death within 30 days of MI was highest in patients with an abnormal LVEF, which in the primary study correlated strongly with QRSd. This conclusion reinforces the belief that QRSd is an indicator of the burden of CVD and probably carries with it an increased risk of SCD post MI. There is some inconsistency in the conclusions of the aforementioned studies examining the risk of SCD post MI with abnormal QRSd. Most of the literature supports an increased risk of all-cause mortality at 30 days, and at longer follow-up, regardless of the method of revascularization used to treat the acute infarction. Why, then, should different conclusions be drawn about the prognostic value of QRSd for SCD events? The probable reason is that there is a good deal of inconsistency between these studies, particularly when the patient population is considered. VALIANT enrolled patients who were post MI with abnormal LVEF, while Bauers study [10] took all comers with MI who were in normal sinus rhythm. The population with abnormal QRSd in the Bauer study [10], felt to be disproportionably sicker than the population with normal intraventricular conduction, had baseline characteristics almost identical to the entire VALIANT population. This suggests that the VALIANT patients with abnormal QRSd have an increased burden of cardiovascular illness and worse ventricular function when compared to Bauers population with abnormal QRSd. This is, in fact, the case and probably serves to identify a population within the VALIANT cohort that is at a higher baseline risk of SCD than the Bauer population. These study population differences act to explain the prognostic value for abnormal QRSd within the VALIANT population and not the Bauer population. Abnormal QRSd is, at least, a risk for all-cause mortality at both 30 days and later, and is also probably prognostic of SCD events, although the data is currently inconclusive. With the prognostic relationship of QRSd post MI probably carrying an increased risk of death and SCD post MI, the next question is whether or not

it should be used in guiding antiarrhythmic therapy during this vulnerable time. The DINAMIT [14] trial primarily examined the efficacy of implantable cardioverter-defibrillator (ICD) placement in patients 640 days post MI with abnormal LVEF, along with impaired cardiac autonomic function (depressed HR variability or elevated 24-hour HR on Holter monitoring), but also stratified the patients by QRSd as a prespecified subgroup. On the whole, the trial did not show any mortality benefit for ICD placement, but it did display a non-significant trend towards benefit for ICD placement in patients with prolongation of the QRS beyond 120 ms. The authors noted that arrhythmic death was significantly decreased, with a similar increase in nonarrhythmic death, resulting in no all-cause mortality benefit for ICD implantation within this population. It is as yet unclear whether prolongation of the QRS complex post acute MI poses an independent risk for SCD. Abnormal QRS duration is prognostic of increased short-term [8, 9] and long-term [5, 10] all-cause mortality probably secondary to the higher burden of CVD within this population. The question of what to do with this information beyond standard post MI aggressive medical therapy remains unanswered.

Ischemic cardiomyopathy
Abnormal intraventricular conduction has been examined extensively as a risk factor for overall mortality, as well as SCD, in patients with ischemic cardiomyopathy (ICM) (Table 2). Earlier studies such as the CASS trial [15] established abnormal intraventricular conduction in patients with ICM as an independent risk factor for all-cause mortality, particularly for LBBB. The limitations of this data are that it was compiled prior to the current revascularization era and did not specifically address arrhythmic death. Current understanding of the relationship between QRS prolongation and risk of SCD in the ICM population has resulted from secondary analyses of multi-center defibrillator databases such as MUSTT and MADIT-II. Although this data is robust, the magnitude of the relationship between abnormal intraventricular conduction and SCD in the ICM population, along with its clinical utility, remains unclear. In their analysis of the MUSTT population, Zimetbaum et al. [16] examined the relationship between numerous ECG abnormalities and allcause mortality/SCD in patients who were not treated with antiarrhythmic therapy (antiarrhythmic

www.cardiologyjournal.org

11

12
Population Dichotomous with 120 ms cut-point SCD in medically treated patients and SCD or VT/VF in ICD group All-cause mortality; median follow-up of 45 months Arrhythmic events in patients followed for more than one year Arrhythmic death QRS variable Primary endpoint Results (95% CI) HR of 2.12 (1.23.76); p = 0.01 in the medical arm and HR of 0.77 (0.471.24); p = 0.28 in the ICD arm HR 0.67 (0.490.93) for 120 ms and 0.84 (0.621.14) for < 120 ms vs placebo in ICD arm Adjusted OR 2.32 (1.375.51) QRS 100 ms; 1.77 (0.973.23) QRS 120 ms; 1.24 (0.652.38) QRS 140 ms IVCD HR 1.44 (1.111.88); p = 0.0069; LBBB HR 1.49 (1.022.17); p = 0.04; RBBB HR 1.05 (0.651.71); p = 0.8383 HR 0.6 (0.30.9) QRS 169 ms; 0.65 (0.350.95) QRS 148168 ms; 0.75 (0.41.25) 147 ms VT/VF treated by ICD in situ VT/VF requiring therapy Log-rank p = 0.016 for QRS 150 ms at two year follow-up OR 1.072 (0.661.74) QRS 120 ms; no QRS duration threshold displayed significant association with arrhythmic events Dichotomous with 120 ms cut-point Continuous 100 ms, 120 ms, 140 ms thresholds Descriptive IVCD, LBBB, RBBB Continuous 147 ms, 148168 ms, 169 ms Dichotomous with 150 ms cut-point Continuous with 120 ms cut-point analysis reported All-cause mortality at 12 months

Table 2. Prognostic significance of QRS duration and morphology in ischemic cardiomyopathy randomized clinical trials.

Trial

Cardiology Journal 2011, Vol. 18, No. 1

MADIT II

1,232

LVEF < 30 one month or more from AMI and three months from revascularization

SCD-HeFT*

2,521 total 1,400 ICM

NYHA class II or III CHF randomized to placebo, amiodarone or ICD

Stecker et al.*

562

Single-centre cohort of ICD recipients for both 1 and 2 prevention

MUSTT

1,638

LVEF 40, no ICD or antiarrhythmic medication and no VT/VF during EPS

www.cardiologyjournal.org

COMPANION*

1,520

NYHA class III or IV CHF randomized to medical therapy, CRT, CRT-ICD

PROFIT*

250

Consecutive patients meeting ACC/AHA criteria for ICD placement (1 and 2 prevention)

PainFREE II

431

CAD ICD for 1 or 2 prevention randomized to ATP vs shock for ventricular arrhythmia

*Includes non-ischemic cardiomyopathy patients in analysis; includes secondary arrhythmia prevention patients in analysis; ICM ischemic cardiomyopathy; LVEF left ventricular ejection fraction; VT ventricular tachycardia; VF ventricular fibrillation; EPS electrophysiological study; CRT cardiac resynchronization therapy; other abbreviations as in Table 1

Andrew Brenyo, Wojciech Zarba, Prognostic significance of QRS duration and morphology

medication or ICD implantation). The ECG abnormalities included: intraventricular conduction delay (IVCD), LBBB, RBBB, left ventricular hypertrophy, and left atrial hypertrophy, with the relevant endpoints being arrhythmic death and all-cause mortality. After multivariate adjustment, only IVCD (HR 1.44), and LBBB (HR 1.49) were associated with arrhythmic death and all-cause mortality. The increase in total mortality associated with LBBB or IVCD, however, outweighed the increase in arrhythmic death risk, suggesting a lack of cause and effect between conduction abnormalities and death. The authors noted that patients with LBBB or IVCD had lower ejection fractions and a higher incidence of symptomatic HF than those without, suggesting that the increase in overall mortality was probably due to a sicker LBBB and IVCD population. These results seem to mesh well with previous data from the CASS registry that displayed increased all-cause mortality associated with LBBB, but not with RBBB, in patients with ischemic heart disease [15]. While the sub-analysis of CASS and MUSTT dealt primarily with medically managed patients with HF, the question of whether or not patients with ICM and abnormal intraventricular conduction were more likely to benefit from ICD placement due to aborted SCD remained. In an initial analysis of both SCD-HEFT and MADIT-II, a greater baseline prolongation of QRS was associated with a trend toward increased benefit from ICD implantation [17, 18]. This association, although present, was not significant in either study. A subsequent analysis of 431 patients in the PainFREE II study did not display any association between QRS duration and ventricular tachycardia (VT)/ventricular fibrillation (VF) in ICD treated patients [19]. Dhar et al. [20] performed an analysis of 731 patients from the MADIT-II cohort with prolonged QRSd, and found it to be an independent predictor of SCD in medically managed patients (HR 2.12) but not in ICD-treated patients (HR 0.77). This is not surprising, because the ICD-treated MADIT II patients died predominantly of non-sudden HF, and QRSd would not be expected to predict HF mortality. At the same time, VT/VF is much more common than SCD, indicating that many ventricular arrhythmias terminate spontaneously. The MADIT-II substudy treated QRSd as a continuous variable, while other studies have utilized an inconsistent cutoff value to determine if QRSd is abnormal. This difference, along with the near identical patient populations (ICD and non-ICD groups), allowed the authors to display a consistent association of lengthened QRSd and SCD across the spectrum of QRSd,

as well as determining the prognostic significance of abnormal intraventricular conduction within these two groups. More recently, QRSd has been used as a stratification tool for arrhythmic risk in patients post ICD placement within the MADIT II cohort [21]. Amongst other basic clinical factors, it was found to have a significant association with events, carrying a HR of 1.56. The model already has been validated, indicating the importance of QRSd among other variables composing MADIT II score [22].

Non-ischemic cardiomyopathy
Compared to the ICM population, there is little data examining the relationship between QRSd and SCD/VF events in the non-ischemic cardiomyopathy (NICM) population (Table 3). Our current understanding of the prognostic implications of abnormal QRSd within the NICM population is the result of subgroup analyses of large multi-center defibrillator efficacy trials, similar to the ICM population. The major exception to this is the Marburg Cardiomyopathy study (MACAS), a kind of natural history cohort study of NICM patients that examined the prognostic implications of baseline patient characteristics in relation to outcomes. Previous small studies had shown that an abnormal signal averaged ECG has prognostic value in predicting ventricular arrhythmias and sudden deaths within the NICM population [23, 24]. However, standard 12lead ECG is the principal source of information regarding QRS duration and morphology. In MACAS [25], a cohort of 343 NICM patients were followed clinically for the development of arrhythmic events with baseline variables obtained including QRSd and morphology. Neither LBBB nor RBBB predicted arrhythmic events within this cohort. The only baseline characteristic that was predictive of VF/SCD events was enrollment LVEF (RR of 2.3 for every 10% decrease in ejection fraction). Although MACAS was observational in nature, it provided excellent information regarding the natural history of NICM and cast doubt upon the predictive utility of QRSd for VF/SCD events. Subsequent to the findings of MACAS, the DEFINITE trial [26] enrolled patients with idiopathic dilated cardiomyopathy and either premature ventricular complexes or non-sustained ventricular tachycardia on 24-hour Holter monitoring and randomized them to either ICD placement or standard medical therapy. Within this population, they found that ICD treatment resulted in an absolute decrease in mortality over standard therapy of 7%

www.cardiologyjournal.org

13

Cardiology Journal 2011, Vol. 18, No. 1

Table 3. Prognostic significance of QRS duration in non-ischemic cardiomyopathy randomized clinical trials.

at two years (14.1 vs 7.9%). This difference was not, however, found to be statistically significant. QRSd was a prespecified subgroup within the study with a QRSd of 120 ms as their dichotomous cutoff point. A trend toward benefit of ICD therapy in patients with a QRSd longer than 120 ms was seen, but again did not reach statistical significance. This should not cause surprise, because the overall DEFINITE trial had limited statistical power and so subgroup analyses were even further underpowered. The shortcomings of this data include the fact that QRSd was treated as a dichotomous variable without specification for morphology or BBB. Previous studies in an ischemic population showed that LBBB patients are at higher risk of arrhythmic death than those with RBBB [15]. Regardless of the shortcomings of this data, it indicates that QRS duration does not have prognostic significance in predicting SCD/VF in the NICM population. SCD-HEFT [17] is the largest randomized clinical trial so far to assess the efficacy of ICD treatment in the NICM population. Its design randomized 1,100 NICM and 1,400 ICM patients to receive amiodarone, ICD or placebo treatment. As a predefined subgroup of this population, the patients were stratified based upon a dichotomous QRS cutoff of 120 ms, but without further stratification into ischemic vs non-ischemic etiologies of LV dysfunction. The group as a whole did display a non-significant trend towards benefit of ICD placement if the QRSd was longer than 120 ms, with the specificity of this result for the NICM population still in question. This finding is consistent with the results of MACAS and the DEFINITE subgroup that did not support abnormal QRSd as having a prognostic role for SCD events. With our current shortcomings regarding risk stratification for SCD in both the ICM and NICM populations, imaging, or more specifically cardiac magnetic resonanse imaging (MRI), has been seen as a possible tool to better identify patients at risk. Hombach et al. [27] examined the utility of cardiac MRI in conjunction with QRSd and QTc in predicting SCD and cardiac death. Of the 149 patients enrolled, 94 had a QRSd of > 110 ms with a statistically significant HR of 5.43 for the primary endpoint. QRSd, along with diabetes, cardiac index and right ventricular end-diastolic volume index on cardiac MRI, was significantly associated with the primary endpoint in Hombachs NICM cohort. Overall, the current data indicates a lack of predictive value of QRSd for SCD events in the NICM population. This conclusion rests on subgroup analysis of ICD efficacy trials either specifically designed for, or that just happened to include,

Multivariate HR for primary endpoint of 1.02 (1.0031.038); p = 0.0008

HR for death QRS 120 ms of 0.5 (0.11.1) favoring ICD therapy; HR for QRS < 120 ms of 0.7 (0.31.5) favoring ICD; difference not significant

HR for SCD of 1.36 for LBBB; p = 0.19 and 1.6 for RBBB; p = 0.57

Multivariate HR of 3.19 (1.47.29); p = 0.045 for QRS > 110 ms Hombach et al. 141 Consecutive patients with NICM meeting WHO criteria who underwent cardiac MR Dichotomous with 110 ms cut-point Cardiac death or SCD with median follow-up of four years

HR 0.67 (0.490.93) for 120 ms and 0.84 (0.621.14) for < 120 ms vs placebo in ICD arm

14

Amiya et al.

SCD-HeFT*

DEFINITE

MACAS

Trial

www.cardiologyjournal.org

*Includes ischemic cardiomyopathy patients in analysis; NICM non-ischemic cardiomyopathy; EF ejection fraction; PVC premature ventricular complexes; NSVT non-sustained ventricular tachycardia; EDD end-diastolic diameter; WHO Word Health Organization; MR magnetic resonanse; other abbreviations as in Table 1

Results (95% CI)

All-cause mortality; median follow-up of 45 months NYHA class II or III CHF randomized to placebo, amiodarone or ICD 2,521 total 1,100 NICM Dichotomous with 120 ms cut-point

SCD with mean follow-up of 29 months

SCD with mean follow-up of 52 months

Dichotomous with 120 ms cut-point

NICM with EF < 45% and EDD of > 56 mm and no prior sustained VT, VF or class IV HF

NICM with EF < 36% and PVCs or NSVT on 24-h Holter monitoring

458

343

78

Consecutive patients with NICM meeting WHO criteria

Population

Dichotomous with 120 ms cut-point

QRS variable

Descriptive LBBB, RBBB

Cardiac death or hospitalization for HF

Primary endpoint

Andrew Brenyo, Wojciech Zarba, Prognostic significance of QRS duration and morphology

patients with NICM. These trials are smaller than those that have examined the same question in the ICM population. No trial involving a NICM population has treated QRSd as a continuous variable, or differentiated patients with QRS prolongation into BBB subgroups. QRSd appears to be a better indicator of the overall burden of CVD, and less of an independent predictor of SCD, within the NICM population.

QRS duration and morphology in cardiac resynchronization therapy patients

Cardiac resynchronization therapy (CRT) is indicated in HF patients with prolonged QRSd ( 120 ms) [2832]. The criterion of wide QRS complex, qualifying HF patients for CRT, indicates that a delay in intraventricular conduction is associated with delayed activation of certain regions of the heart and impaired dysynchronous contraction. Applying LV or biventricular pacing in such patients improves electrical activation of the ventricles and improves contractility by synchronizing movements of the LV free wall and interventricular septum. Previous studies have documented that prolonged QRSd, especially if associated with a LBBB, is associated with more advanced LV dysfunction in HF patients. However, the prognostic significance of BBB in HF patients remains controversial. Baldasseroni et al. [33] evaluated the prognostic significance of QRS morphology in 5,517 HF outpatients, with LBBB proving to be predictive of all-cause mortality when compared to patients without LBBB. Barsheshet et al. [34] evaluated 4,102 hospitalized HF patients. RBBB, but not LBBB, was found to be predictive for mortality. Differences in clinical characteristics of studied patients probably contribute to these differences. The prognostic significance of QRS morphology has been evaluated in several CRT trials. In the Comparison of Medical Therapy, Pacing, and Defibrillation in Heart Failure (COMPANION) trial [28] enrolling class III and IV HF patients, there was no significant difference in the risk of primary or secondary endpoints between patients with LBBB and patients with RBBB or IVCD. In the study by Iler et al. [35] of 337 CRT patients, LBBB was not predictive of death or heart transplant outcome. In another cohort of 636 CRT patients [36], three-year average survival was significantly better in LBBB than in RBBB patients. In a study by Wokhlu et al. [37] of 505 CRT patients, survival over a median

2.6 years of follow-up was significantly better in LBBB than in RBBB patients receiving CRT. These observations indicate that a therapy, in this case CRT, might modify the prognostic significance of QRS morphology: LBBB patients seem to benefit more from CRT than RBBB patients, and therefore have better survival. The MADIT-CRT [38] (Multicenter Automatic Defibrillator Implantation Trial-Cardiac Resynchronization Therapy) evaluated the effects of CRT-D vs ICD-only therapy in 1,820 mild to moderate HF patients with ischemic (NYHA class I or II) and nonischemic (class II) cardiomyopathy with ejection fraction 30% and QRS 130 ms. The CRT-D therapy was associated with a 34% reduction in the risk of primary endpoint, defined as death from any cause or a non-fatal HF event (whichever came first). Patients with QRS duration 150 ms derived significantly more benefit from CRT-D than did patients with QRS < 150 ms. The LBBB patients demonstrated the most significant effect of resynchronization therapy in terms of reduction of HF progression, whereas no such benefit was observed in the RBBB and IVCD patients [39]. The MADIT-CRT trial provided evidence that CRT very dramatically reduces the progression of HF in relatively asymptomatic or mildly symptomatic patients with a low ejection fraction and wide QRS complex. Similarly, in recently reported results of extended follow-up from the REVERSE trial [40] that studied patients with HF in NYHA class I or II with QRS 120 ms, the authors found that patients with QRSd 152 ms tended to derive more benefit from CRT (measured by composite HF status) than patients with a QRS shorter than 152 ms. The results of the Resynchronization/Defibrillation in Advance Heart Failure Trial (RAFT) provide further evidence for the importance of wide QRS complex and LBBB morphology [41]. RAFT enrolled 1,798 NYHA class II or III HF patients with an ejection fraction 30% and an intrinsic QRS 120 ms or a paced QRS 200 ms, and randomized them to an ICD or CRT-D. The risk of primary endpoint (HF hospitalization or death) was significantly reduced, by 25% (HR 0.75; p < 0.001). Patients with wide QRS benefited from CRT-D therapy significantly more than patients with QRS < 150 ms and patients with LBBB benefited more from CRT-D than patients with RBBB, IVCD or paced QRS. RAFT also showed a significant decrease in mortality, which was significantly higher than in MADIT-CRT [42].

www.cardiologyjournal.org

15

Cardiology Journal 2011, Vol. 18, No. 1

Summary
QRS duration and morphology should be considered as important prognostic information indicating more advanced cardiac pathology secondary to underlying heart disease, whether that heart disease is known about or not. Post-infarction patients with prolonged QRSd have a significantly increased risk of mortality, although data associating QRS prolongation specifically with sudden death is less supportive. In NICM, there is no evidence that QRSd carries prognostic significance for predicting mortality or sudden death. However, cohorts of patients with HF and low ejection fraction usually include about 50% of patients with NICM; in such studies, a significantly prolonged QRSd, and especially the presence of LBBB, seem to predict benefit from CRT-D therapy in both ICM and NICM patients [43]. Therefore, QRS duration and morphology should not only be considered as a predictor of death or sudden death, but also as a predictor of benefit from CRT in patients with HF, whether of ischemic or non-ischemic origin.

8. Hathaway WR, Peterson ED, Wagner GS et al. Prognostic significance of the initial electrocardiogram in patients with acute myocardial infarction. GUSTO-I Investigators. Global Utilization of Streptokinase and t-PA for Occluded Coronary Arteries. JAMA. 1998; 279: 387391. 9. Wong CK, Gao W, Stewart RA, French JK, Aylward PE, White HD. Relationship of QRS duration at baseline and changes over 60 min after fibrinolysis to 30-day mortality with different locations of ST elevation myocardial infarction: Results from the Hirulog and Early Reperfusion or Occlusion-2 trial. Heart, 2009; 95: 276282. 10. Bauer A, Watanabe MA, Barthel P, Schneider R, Ulm K, Schmidt G. QRS duration and late mortality in unselected post-infarction patients of the revascularization era. Eur Heart J, 2006; 27: 427433. 11. Guerrero M, Harjai K, Stone GW et al. Comparison of the prognostic effect of left versus right versus no bundle branch block on presenting electrocardiogram in acute myocardial infarction patients treated with primary angioplasty in the primary angioplasty in myocardial infarction trials. Am J Cardiol, 2005; 96: 482488. 12. Yerra L, Anavekar N, Skali H et al. Association of QRS duration and outcomes after myocardial infarction: The VALIANT trial. Heart Rhythm, 2006; 3: 313316. 13. Solomon SD, Zelenkofske S, McMurray JJ et al. Sudden death in patients with myocardial infarction and left ventricular dysfunction, heart failure, or both. N Engl J Med, 2005; 352: 25812588. 14. Hohnloser SH, Kuck KH, Dorian P et al. Prophylactic use of an implantable cardioverter-defibrillator after acute myocardial infarction. N Engl J Med, 2004; 351: 24812488. 15. Coronary artery surgery study (CASS): A randomized trial of coronary artery bypass surgery. Survival data. Circulation, 1983; 68: 939950. 16. Zimetbaum PJ, Buxton AE, Batsford W et al. Electrocardiographic predictors of arrhythmic death and total mortality in the multicenter unsustained tachycardia trial. Circulation, 2004; 110: 766769. 17. Bardy GH, Lee KL, Mark DB et al. Amiodarone or an implantable cardioverter-defibrillator for congestive heart failure. N Engl J Med, 2005; 352: 225237. 18. Moss AJ, Zareba W, Hall WJ et al. Prophylactic implantation of a defibrillator in patients with myocardial infarction and reduced ejection fraction. N Engl J Med, 2002; 346: 877883. 19. Buxton AE, Sweeney MO, Wathen MS et al. QRS duration does not predict occurrence of ventricular tachyarrhythmias in patients with implanted cardioverter-defibrillators. J Am Coll Cardiol, 2005; 46: 310316. 20. Dhar R, Alsheikh-Ali AA, Estes NA 3rd et al. Association of prolonged QRS duration with ventricular tachyarrhythmias and sudden cardiac death in the Multicenter Automatic Defibrillator Implantation Trial II (MADIT-II). Heart Rhythm, 2008; 5: 807813. 21. Goldenberg I, Vyas AK, Hall WJ et al. Risk stratification for primary implantation of a cardioverter-defibrillator in patients with ischemic left ventricular dysfunction. J Am Coll Cardiol, 2008; 51: 288296. 22. Garikipati NV PV, Adhaduk M et al. Risk stratification of ICD patients with ischemic cardiomyopathy: Do clinical trial results apply in the real world? Circulation, 2010; A16147. 23. Poll DS, Marchlinski FE, Falcone RA, Josephson ME, Simson MB. Abnormal signal-averaged electrocardiograms in patients with nonischemic congestive cardiomyopathy: Relationship to sus-

Acknowledgements
The authors do not report any conflict of interest regarding this work.

References
1. Rabkin SW, Mathewson FL, Tate RB. The electrocardiogram in apparently healthy men and the risk of sudden death. Br Heart J, 1982; 47: 546552. Schinkel AF, Elhendy A, van Domburg RT et al. Prognostic significance of QRS duration in patients with suspected coronary artery disease referred for noninvasive evaluation of myocardial ischemia. Am J Cardiol, 2009; 104: 14901493. Elhendy A, Hammill SC, Mahoney DW, Pellikka PA. Relation of QRS duration on the surface 12-lead electrocardiogram with mortality in patients with known or suspected coronary artery disease. Am J Cardiol, 2005; 96: 10821088. Petrina M, Goodman SG, Eagle KA. The 12-lead electrocardiogram as a predictive tool of mortality after acute myocardial infarction: Current status in an era of revascularization and reperfusion. Am Heart J, 2006; 152: 1118. Brilakis ES, Wright RS, Kopecky SL, Reeder GS, Williams BA, Miller WL. Bundle branch block as a predictor of long-term survival after acute myocardial infarction. Am J Cardiol, 2001; 88: 205209. Hindman MC, Wagner GS, JaRo M et al. The clinical significance of bundle branch block complicating acute myocardial infarction. 1. Clinical characteristics, hospital mortality, and one-year follow-up. Circulation, 1978; 58: 679688. Nimetz AA, Shubrooks SJ Jr, Hutter AM Jr, DeSanctis RW. The significance of bundle branch block during acute myocardial infarction. Am Heart J, 1975; 90: 439444.

2.

3.

4.

5.

6.

7.

16

www.cardiologyjournal.org

Andrew Brenyo, Wojciech Zarba, Prognostic significance of QRS duration and morphology

tained ventricular tachyarrhythmias. Circulation, 1985; 72: 13081313. 24. Mancini DM, Wong KL, Simson MB. Prognostic value of an abnormal signal-averaged electrocardiogram in patients with nonischemic congestive cardiomyopathy. Circulation, 1993; 87: 10831092. 25. Grimm W, Christ M, Bach J, Muller HH, Maisch B. Noninvasive arrhythmia risk stratification in idiopathic dilated cardiomyopathy: Results of the Marburg Cardiomyopathy Study. Circulation, 2003;108: 28832891. 26. Kadish A, Dyer A, Daubert JP et al. Prophylactic defibrillator implantation in patients with nonischemic dilated cardiomyopathy. N Engl J Med, 2004; 350: 21512158. 27. Hombach V, Merkle N, Torzewski J et al. Electrocardiographic and cardiac magnetic resonance imaging parameters as predictors of a worse outcome in patients with idiopathic dilated cardiomyopathy. Eur Heart J, 2009; 30: 20112018. 28. Bristow MR, Saxon LA, Boehmer J et al. Cardiac-resynchronization therapy with or without an implantable defibrillator in advanced chronic heart failure. N Engl J Med, 2004; 350: 21402150. 29. Cleland JG, Daubert JC, Erdmann E et al. The effect of cardiac resynchronization on morbidity and mortality in heart failure. N Engl J Med, 2005; 352: 15391549. 30. Cazeau S, Leclercq C, Lavergne T et al. Effects of multisite biventricular pacing in patients with heart failure and intraventricular conduction delay. N Engl J Med, 2001; 344: 873880. 31. Jeevanantham V, Daubert JP, Zareba W. Cardiac resynchronization therapy in heart failure patients: An update. Cardiol J, 2009; 16: 197209. 32. Zareba W, Zareba KM. Which patients with chronic heart failure should be referred for CRT-D implantation? Practical implications of current clinical research. Pol Arch Med Wewn, 2010; 120: 95102. 33. Baldasseroni S, Opasich C, Gorini M et al. Left bundle-branch block is associated with increased one-year sudden and total mortality rate in 5,517 outpatients with congestive heart failure: A report from the Italian network on congestive heart failure. Am Heart J, 2002; 143: 398405.

34. Barsheshet A, Leor J, Goldbourt U et al. Effect of bundle branch block patterns on mortality in hospitalized patients with heart failure. Am J Cardiol, 2008; 101: 13031308. 35. Iler MA, Hu T, Ayyagari S et al. Prognostic value of electrocardiographic measurements before and after cardiac resynchronization device implantation in patients with heart failure due to ischemic or nonischemic cardiomyopathy. Am J Cardiol, 2008; 101: 359363. 36. Adelstein EC, Saba S. Usefulness of baseline electrocardiographic QRS complex pattern to predict response to cardiac resynchronization. Am J Cardiol, 2009; 103: 238242. 37. Wokhlu A, Rea RF, Asirvatham SJ et al. Upgrade and de novo cardiac resynchronization therapy: Impact of paced or intrinsic QRS morphology on outcomes and survival. Heart Rhythm, 2009; 6: 14391447. 38. Moss AJ, Hall WJ, Cannom DS et al. Cardiac-resynchronization therapy for the prevention of heart-failure events. N Engl J Med., 2009; 361: 13291338. 39. Zareba W KH, Cygankiewicz I et al. Effectiveness of cardiac resynchronization therapy by QRS morphology in the Multicenter Automatic Defibrillator Implantation Trial-Cardiac Resynchronization Therapy (MADIT-CRT). Circulation, 2011 (in press). 40. Linde C, Abraham WT, Gold MR, St John Sutton M, Ghio S, Daubert C. Randomized trial of cardiac resynchronization in mildly symptomatic heart failure patients and in asymptomatic patients with left ventricular dysfunction and previous heart failure symptoms. J Am Coll Cardiol, 2008; 52: 18341843. 41. Tang AS, Wells GA, Talajic M et al. Cardiac-resynchronization therapy for mild-to-moderate heart failure. N Engl J Med, 2010; 363: 23852395. 42. Zarba W. Comparison of clinical trials evaluating cardiac resynchronization therapy in mild to moderate heart failure. Cardiol J, 2010; 17: 543548. 43. Barsheshet A GI, Moss AJ et al. Response to preventive cardiac resynchronization therapy in patients with ischaemic and nonischaemic cardiomyopathy in MADIT-CRT. Eur Heart J, 2011 (in press).

www.cardiologyjournal.org

17